international journal of marketing and technology (issn ... doc/ijmt_september/ijmra-mt338.pdf · 5...

International Journal of Marketing and Technology (ISSN: 2249-1058)

CONTENTS

Sr.

No. TITLE & NAME OF THE AUTHOR (S) Page

No.

1 Comparative study of International vs. Traditional HRM Issues and Challenges.

Sirous Fakhimi-Azar, Farhad Nezhad Haji Ali Irani and Mohammad Reza Noruzi 1-16

2 An Analytical Study of Marketing of Banking Services of SBI and HDFC Bank in Borivali Suburb,

Mumbai.

Dr. M. N. Sondge and Prof. T. B. Gadhave

17-39

3 Investor’s Awareness and Preference Towards Mutual Funds Investments - Some Survey Evidences.

Dr. Megha Sandeep Somani 40-61

4 The Implication of Moral Intelligence and Effectiveness in Organization; Are They Interrelated?

Gholam Reza Rahimi 62-76

5 Magnitude and Compensability of Industrial Accidents in Nepal.

Dr. Shyam Bahadur Katuwal 77-100

6 Marketing of Dwcra Products A New Pardigm for Combating Rural Poverty - A Case Study Of

Andhra Pradesh.

Dr. K. Lalith and Prof. G. Prasad

101-111

7 Analyzed Traffic Through Switches In The Design of LANs using OPNET MODELER.

Mr. Ishu Gupta, Dr. Harsh Sadawarti and Dr. S. N. Panda 112-124

8 Customer Satisfaction of Retail Consumers With Special Relevance To Organized Retail Outlets In

Chennai City.

Anita Priscilla .J and Dr. Shanthi

125-145

9 Cluster Based Mutation Testing Using Homogeneous and Heterogeneous N-MUTANTS.

Mr. Ajay Jangra and Ms. Jasleen kaur 146-160

10 Review of Supply Chain Management for Modeling and Integration in Indian Electronics &

Telecommunication industry.

Parul Goyal

161-181

11 Issues and Perspectives on Two fundamental Intangible Assets in Organizations; Intellectual and

Social Capitals.

Firouze Azizi and Mohammad Reza Noruzi

182-197

12 Management of Transportation System and Prioritization of Transport Infrastructure Projects.

Jayanti De, Dr. Sudip Kumar Roy and Dr. Madhumati Dutta 198-214

13 Mobile Learning Empowering Rural Women A study of Vidiyal (NGO) in Theni District,

TAMILNADU.

Dr. (Mrs.) S. Hasan Banu

215-243

14 A study on point of purchase - An Advertising and Selling Technique.

Mrs. Priti Jeevan 244-263

IJMT Volume 1, Issue 4 ISSN: 2249-1058 __________________________________________________________

A Monthly Double-Blind Peer Reviewed Refereed Open Access International e-Journal - Included in the International Serial Directories Indexed & Listed at: Ulrich's Periodicals Directory ©, U.S.A., Open J-Gage as well as in Cabell’s Directories of Publishing Opportunities, U.S.A.

International Journal of Marketing and Technology http://www.ijmra.us

147

September

2011

Chief Patron Dr. JOSE G. VARGAS-HERNANDEZ

Member of the National System of Researchers, Mexico

Research professor at University Center of Economic and Managerial Sciences,

University of Guadalajara

Director of Mass Media at Ayuntamiento de Cd. Guzman

Ex. director of Centro de Capacitacion y Adiestramiento

Patron Dr. Mohammad Reza Noruzi

PhD: Public Administration, Public Sector Policy Making Management,

Tarbiat Modarres University, Tehran, Iran

Faculty of Economics and Management, Tarbiat Modarres University, Tehran, Iran

Young Researchers' Club Member, Islamic Azad University, Bonab, Iran

Editorial Board

Dr. CRAIG E. REESE Professor, School of Business, St. Thomas University, Miami Gardens

Dr. S. N. TAKALIKAR Principal, St. Johns Institute of Engineering, PALGHAR (M.S.)

Dr. RAMPRATAP SINGH Professor, Bangalore Institute of International Management, KARNATAKA

Dr. P. MALYADRI Principal, Government Degree College, Osmania University, TANDUR

Dr. Y. LOKESWARA CHOUDARY Asst. Professor Cum, SRM B-School, SRM University, CHENNAI

Prof. Dr. TEKI SURAYYA Professor, Adikavi Nannaya University, ANDHRA PRADESH, INDIA




148

September

2011

Dr. T. DULABABU Principal, The Oxford College of Business Management, BANGALORE

Dr. A. ARUL LAWRENCE SELVAKUMAR Professor, Adhiparasakthi Engineering College, MELMARAVATHUR, TN

Dr. S. D. SURYAWANSHI Lecturer, College of Engineering Pune, SHIVAJINAGAR

Dr. S. KALIYAMOORTHY Professor & Director, Alagappa Institute of Management, KARAIKUDI

Prof S. R. BADRINARAYAN Sinhgad Institute for Management & Computer Applications, PUNE

Mr. GURSEL ILIPINAR ESADE Business School, Department of Marketing, SPAIN

Mr. ZEESHAN AHMED Software Research Eng, Department of Bioinformatics, GERMANY

Mr. SANJAY ASATI

Dept of ME, M. Patel Institute of Engg. & Tech., GONDIA(M.S.)

Mr. G. Y. KUDALE

N.M.D. College of Management and Research, GONDIA(M.S.)

Editorial Advisory Board

Dr.MANJIT DAS Assistant Professor, Deptt. of Economics, M.C.College, ASSAM

Dr. ROLI PRADHAN Maulana Azad National Institute of Technology, BHOPAL




149

September

2011

Dr. N. KAVITHA Assistant Professor, Department of Management, Mekelle University, ETHIOPIA

Prof C. M. MARAN Assistant Professor (Senior), VIT Business School, TAMIL NADU

DR. RAJIV KHOSLA Associate Professor and Head, Chandigarh Business School, MOHALI

Dr. S. K. SINGH Asst. Professor, R. D. Foundation Group of Institutions, MODINAGAR

Dr. (Mrs.) MANISHA N. PALIWAL Associate Professor, Sinhgad Institute of Management, PUNE

DR. (Mrs.) ARCHANA ARJUN GHATULE Director, SPSPM, SKN Sinhgad Business School, MAHARASHTRA

DR. NEELAM RANI DHANDA Associate Professor, Department of Commerce, kuk, HARYANA

Dr. FARAH NAAZ GAURI Associate Professor, Department of Commerce, Dr. Babasaheb Ambedkar Marathwada

University, AURANGABAD

Prof. Dr. BADAR ALAM IQBAL Associate Professor, Department of Commerce, Aligarh Muslim University, UP

Associate Editors

Dr. SANJAY J. BHAYANI Associate Professor ,Department of Business Management, RAJKOT (INDIA)

MOID UDDIN AHMAD Assistant Professor, Jaipuria Institute of Management, NOIDA

Dr. SUNEEL ARORA Assistant Professor, G D Goenka World Institute, Lancaster University, NEW DELHI




150

September

2011

Mr. P. PRABHU Assistant Professor, Alagappa University, KARAIKUDI

Mr. MANISH KUMAR Assistant Professor, DBIT, Deptt. Of MBA, DEHRADUN

Mrs. BABITA VERMA Assistant Professor, Bhilai Institute Of Technology, DURG

Ms. MONIKA BHATNAGAR Assistant Professor, Technocrat Institute of Technology, BHOPAL

Ms. SUPRIYA RAHEJA Assistant Professor, CSE Department of ITM University, GURGAON




151

September

2011

CLUSTER BASED MUTATION TESTING USING

HOMOGENEOUS AND HETEROGENEOUS

N-MUTANTS

Mr. Ajay Jangra

CSE Department

U.I.E.T. Kurukshetra University,

Kurukshetra, INDIA

Ms. Jasleen kaur

CSE Department

U.I.E.T. Kurukshetra University,

Kurukshetra, INDIA

Title

Author(s)




152

September

2011

ABSTRACT:

Mutation testing is to check the quality of test cases by changing code. The proposal examines

mutation testing based on clusters.Clusters contains homogeneous and heterogeneous mutants. It

consists two mutagens in it and by one time only one mutagen (mutant) can be selected for

performing mutation testing. Mutants can be homogeneous type or it can be heterogeneous

mutants. Selection of number of mutants depends upon number of clusters working for the code.

This paper represents cluster based testing using homogeneous and heterogeneous mutants. This

is divided into three sections. Section 1, Section 2 and Section 3 described about homogeneous,

heterogeneous and hybrid (homogeneous/ heterogeneous) mutants. For mutation generation, a

tester can generate mutants according to number of clusters. Number of cluster is equivalent to

number of mutants. After modifying original code, collaboration of all mutants performed which

shows by end a tester can perform mutation on huge number of mutants. This paper explores the

ideas of collaboration of cluster covers large number of mutants.

Keywords: Mutant, Mutation Operator, Equivalent Mutant, Test cases, Clusters.

INTRODUCTION:

Mutation testing (fault based testing) checks the quality of test cases by feeding wrong data into

original code. Changes in original code makes mutant in program called as mutated program. It

works on single mutant for checking the efficiency and effectiveness of test cases. Mutation

testing is white box testing tests the internal structure of the code to detect all the mistakes in real

code. Figure 1 represents the complete scenarios of the mutation process [1, 2].

Figure: 1-Mutation process




153

September

2011

A correct program can be shown in many wrong ways so mutation testing work under this

hypotheses by making correct program to incorrect forms by using fault seeding concept. Fault

seeding is done by using different kinds of mutation operators. Mutation operators are the basic

and simple rules in mutation testing for the transformation of the original code in many other

various mirror images with little changes in them. Mutation operators change one operand to

another operand or may delete the entire statement [3, 4, 5].

PREVIOUS WORK:

Work done on mutation testing is like some syntactic changes in the procedural and

object oriented languages, types of mutation operators were introduced and different tools

were used to show the mutation testing [6, 8].

Two types of mutation testing are described in paper “Weak mutation testing and

completeness of test sets” that one is strong mutation testing creates the mutant for the

whole program and another one is weak mutation testing creates the mutant in the single

statement. If the output shows the same result as the original program then that test case

is not either effective or efficient [4,7].

Different classification of mutation operators has defined i.e. method level mutation

operators and class level mutation operators where the method level deals with

arithmetic syntactic changes and class level deals with the combination of non object

oriented programming and object oriented programming language in which the program

contains polymorphism, inheritance, templates and exceptional handling [11,12, 13].

PROBLEM DOMAIN:

Expensive due to generation large number of mutant[9]

Equivalent mutant’s execution gives the same result and it’s very difficult to detect. It

also shows the disadvantage in terms of cost and time [10].

Development of mutagen for different code increases effort and time [5].




154

September

2011

PROPOSED MODEL:

Testing approach called fault based testing is presented for overcoming the drawbacks of other

testing approaches. However it seems to be expensive due to generation of single mutagen

(mutant) version copies for the original code. Creation of different mutated codes for checking

original one is time consuming and needs effort for testing. Mutation testing provides the clean

way to achieve the goal. Clean way means effort can put in the early stages so that the less effort

or time will be at last

Proposed model develops “Cbm-nm (Cluster based mutation testing using homogeneous and

heterogeneous n-mutant” can be considered as already built in for testing. It means clusters are

fixed and number of clusters depends upon the tester and the code for performing testing.

Theoretical way of defining Cluster based model is presented here. Testing implementation

depends upon the cluster. So a cluster plays an important role for coverage of n-mutants.

User can choose only a mutant at a time from cluster. So if number of cluster is four then user

can pick one from each and can do coverage of 4 mutants at a time. Each cluster consists two

mutants or double mutants in them. That’s why the proposed model work can work on n-

mutants.

Double mutants can be created by two ways such as:-

1. Homogeneous mutants: Homogeneous mutants are the same type of operators mistakes

such as (+, +), (-,-), (*,*), (/, /), (%,%).

2. Heterogeneous mutants: Heterogeneous mutants are the different type of mutation

operators such as (+, -), (+,-), (+,*), (+, /), (+, %).

Figure-3 Collaboration of clusters




155

September

2011

Figure 3 shows after generating mutants, cluster formation makes pairing with another cluster.

By collaborating one cluster to another tester can pick two mutants at a time and the pairing three

clusters tester can pick three mutants and so on. Therefore number of mutants depends upon

number of clusters. So by the end, there is integration of all clusters in one cluster which can

cover n-mutants in single execution which shows high coverage of mutants.

There are 3 sections for performing mutation testing.

Section 1: Homogeneous mutant’s coverage with values a=9, b=8, c=7, d=9.

Section 2: Heterogeneous mutant’s coverage with values a=9, b=8, c=7, d=9.

Section 3: Hybrid mutant’s coverage (Homogeneous/ heterogeneous mutants).

Figure 3 Original code to be tested

Figure-4 generates homogeneous, heterogeneous and hybrid cluster for mutation. Cluster 1 &

cluster 2 is for section 1, Cluster 3 & cluster 4 is for section 2 and Cluster 5 & cluster 6 is for

section 3.

Figure 4- formation of clusters




156

September

2011

How mutation testing do coverage with these clusters is shown below:-

Section 1: Select m1 from c1 and m2 from c2 and do checking of dual mutants in single

execution. It gives different result b=8, which shows high coverage and develop highly efficient

test case.

Figure 5- checking on dual mutants


execution. It gives different result c=8 reveal high coverage and develop highly efficient test

case.


execution. It gives b=8 different output from original. Hence develop highly efficient test case.

RESULTS:

Formation of different clusters within a single code and combining them later shows higher

coverage of mutants. So heterogeneous and homogeneous concept shows only the huge coverage

in them for performing mutation testing in which clusters are important to define.

0

50

100

150

probabilty

cluster c1cluster c1,2cluster c1,2,3cluster c1,2,3,4cluster c1,2,3,4,5cluster c1,2,3,4,5,6




157

September

2011

Figure 6 Probability determination

Figure 6: explores that combination of all mutants shows the high probability of coverage of

mutants which can generate large mutants and can generate highly efficient and effective test

case for testing.

Figure 7: Hierarchal representations of clusters

Graph 7 depicts that as the cluster moves downwards by attaching with each other then number

of mutants also increases which means that last collaboration shows the higher large number of

mutants.

Number of clusters α number of mutants

Probability of clusters mutants = number of mutants chosen/ total number of cluster

Maximum probability can be 1 which means c1+c2+c3+c4 =1.

Maximum probability of coverage can be considered as 1.

Figure 8: Mutants coverage against clusters

So mutant’s execution can be done in single execution at last cluster combination which shows

in figure 8.

MU

TAN

TS C

OV

ERA

GE

Mutants

Coverage




158

September

2011

Figure 9 shows benefits of proposal work. It increases the effectiveness and efficiency of the test

cases and decreases the time, cost and effort. This can be considered as future scope of the

dissertation. By making automatic framework for selecting one mutant from each pre-defined

clusters and executing those meta mutants by single execution which never generate the bad test

case or failed test case.

Figure 9: Performance

MAJOR ACHIEVEMENTS:

The proposal reveals the following achievements:-

Quality of test cases increases.

Effort gets reduced.

Budget and time also decreases because the n-mutant worked efficiently.

APPLICATION AREAS:

Mutation testing can be useful in software engineering as well as in medical field.

To check the quality of test cases, mutation testing can be applied to software’s.

Automatic generation of effective and efficient test cases.

It can be used to check the circuit system as well by changing the behavior of the circuit,

by removing the functionality of the circuit or by inserting nee functions in to the circuit.

EffectivenessEfficiency

Time EffortCost

M

U

T

A

N

T

I

N

C

R

E

A

S

E

S

performance Effectiveness

Efficiency




159

September

2011

It can also be used in medical purpose like for DNA checking. Each gene of the human

being contains chromosomes which can be test by applying mutation into those

chromosomes in a different manner.

Figure-10: DNA structure testing

By applying mutation testing in the form of (0,0), (0,1), (1,0),(1,1). There can be probability of

2ⁿ cases for each cluster. Suppose the single DNA part contains the four parts has probability of

2ⁿ where n=4 that is 16 cases.

CONCLUSION:

Mutation testing analysis with homogeneous and heterogeneous mutants describes checking of

n-mutants. By getting through all the necessary details of the mutation testing, result shows great

benefit in the problem area of cost and time. Till today the authors described about single mutant

and this paper reveals homogeneous and heterogeneous mutants testing concept. So instead of

creating single mutagen, n-mutants can be generated with predefined clusters which can reduce

time and cost instead of that it shows the great benefit in the area of mutation testing.

REFERENCES:

WILLIAM E. HOWDEN “Weak Mutation Testing and Completeness Of Test Sets” IEEE

1985.

A. Jefferson Offutt “Investigations of the Software testing Coupling Effect” Clemson

University.

T. A. Budd. Mutation analysis of program test data. PhD thesis, Yale University, New

Haven CT , USA, 1980.




160

September

2011

T.H.Tse, T.Y. Chen, X.Feng “On the completeness of Test Cases for Atomic Arithmetic

Expressions” IEEE 2000.

Huo Yan Chen, Su Hu “Two new kinds of class level mutants for object oriented programs”

IEEE 2006.

Yue Jia, Mark Harman "An Analysis and Survey of the Development of Mutation Testing"

King's College London, Technical Report.

Hiralal Agarwal, Richard A. DeMillo, Bob Hathaway, William Hsu, E.w. Krauser, R.J.

Martin, Aditya P. Mathur, Eugene Spaord “Design of mutant operators for C programming

language” 2006.

Laurie Williams “Mutation testing” Research group software engineering in NCSU 2007.

M. R. Woodward “Mutation testing—its origin and evolution” Department of Computer

Science, Chadwick Building, University of Liverpool, PO Box 147, Liverpool L69 3BX, UK.

John A. Clark and Haitao Dan and Robert M. Hierons “Semantic Mutation Testing”

Proceedings of the 5th International Workshop on Mutation Analysis

(MUTATION'10)Paris, France, 6 April 2010.

Mrs. R. Jeevarathinam, Dr.Antony Selvadoss Thanamani “Test Case Generation using

Mutation Operators and Fault Classification” 2010.

L. Zhang, S.-S. Hou, J.-J Hu, T. Xie and H. mei “Is Operator-Based Mutant Selection

Superior to Random Mutant Selection?” in Proc. ICSE, 2010.

Ajay Jangra,

Jasleen Kaur “Operator Based Efficient Mutant Generation Methods for

Procedural and Object-Oriented Languages” IJCST vol2 issue2, 2011.

http://www.dcs.kcl.ac.uk/pg/jiayue/repository/TR-09-06.pdf

international journal of marketing and technology (issn ... doc/ijmt_september/ijmra-mt338.pdf · 5...

Documents